While substantial progress has been made in prevention, detection, and treatment of certain forms of cancer, it remains one of the leading causes of death in the developed world. Certain forms of cancer, such as prostate cancer, have high 5-year survival rates when detected in the early stages, but much lower survival rates when detected after metastasis. There is thus a need for new and improved agents for treating cancer and other diseases marked by abnormal cell proliferation.
Polyamines are naturally-occurring compounds which are essential for the growth and division of cells. Polyamines (including putrescine, spermidine, and spermine) increase in proliferating issues. A number of polyamine analogues have shown promise as anticancer agents. They are able to kill cells and inhibit cell growth both in vitro and in vivo. Most successful among these analogues have been the α-N, ω-N alkyl derivatives of the higher and lower homologues of spermine, although several alkylated diamines also show promise as inhibitors of tumor cell proliferation. See, e.g., U.S. Pat. Nos. 5,541,230 and 5,880,161. Many hypotheses have been advanced to explain the biological effects of the polyamines; one of the most compelling hypotheses attributes their effects to polyamine binding to nucleic acids and to receptor targets. Since spermidine and spermine are strong bases, they are protonated at physiological pH and can therefore bind to the negatively charged nucleic acids either by electrostatic interactions or by hydrogen bonding.
Polyamines are known to interact with and induce structural changes in DNA in cell-free systems. Spermidine and spermine can cause DNA to condense and aggregate and can induce both B-to-Z and B-to-A transitions in certain DNA sequences. Molecular mechanics studies of spermine-DNA interactions have shown that, in a minimum energy conformation, spermine is bound in a cisoidal conformation that wraps around the major groove of the double helix. Spermine and its α-N, ω-N-bisethylated higher and lower homologues, as well as spermidine, have been shown to bind to t-RNA. Using 1H NMR analysis, spermine and 1N, 12N-bisethylspermine (BES) were found to bind at the TψC loop of t-RNAPhe. The binding is not of an electrostatic nature, but rather a consequence of the different hydrogen-bonding modes that can be established between both types of molecules. Finally, polyamines have recently been shown to bind to receptors such as the N-methyl-D-glutamate (NMDA) receptor and the glutamate receptor (Glu-R) and to block and modulate a number of ion channels, results that open new vistas in the pharmacology of the polyamines.
DNA-interacting drugs have long been of interest as anticancer agents. Mechanical models of the DNA double helix have created an image of a rigid structure; however, experimental evidence suggests that DNA has considerable flexibility. When designing polyamine analogues that can bind to DNA, structural modifications considered include variations in the number of distance of carbons between nitrogens and/or terminal N-substitutions of different types. This invention is directed to the design and synthesis of chiral analogues of spermine and other naturally and non-naturally occurring polyamines that bind DNA, t-RNA, or other polyamine binding sites, in a modified and selective fashion due to their increased conformational rigidity. The introduction of restriction in the free rotation about the single bonds in a flexible molecule such as spermine (which has a myriad of potential conformations) can result in spatial rigidity which can introduce bends, kinks, or loops at their binding domains. The introduction of conformational restriction has been very fruitful in the design of peptidomimetics, and a recent report described the synthesis of chiral pyrrolidyl polyamines.
As a starting point, rigid analogues of spermine were constructed where the added atoms or bonds had minimal effect on the size and molecular weight of the parent compound. The simple addition of a cyclopropyl ring to the butane segment of spermine introduces chirality and conformational restriction in an otherwise flexible molecule.

The structure (I) shows the conformation of 1N, 12N-bisethylspermine (BES), while the two structures (II) and (III) show the conformation of the trans-isomer (II) and cis-isomer (III) resulting from replacing the central butane segment in BES with trans- and cis-1,2-dimethylcyclopropyl residues, respectively.
Cyclopropane derivatives are also known to have important biological functions. Incorporating a cyclopropane moiety into a polyamine structure can enhance the anticancer effect of the polyamine analog via additional mechanisms complementary to the conformational effect on the polyamine analog.
The design of new and more efficacious tetramines offers several advantages. While the precise mechanisms by which polyamines kill tumor cells and cause systemic toxicity are still not entirely clear, it has been established that polyamines and their analogues bind to nucleic acids and alter their conformations, that they bind to receptor targets, that they drastically reduce the level of ornithine decarboxylase (the first enzyme in the pathway that leads to spermine biosynthesis in mammals), that they upregulate the levels of spermidine/spermine N-acetyltransferase (the enzyme involved in the catabolism and salvage pathways of spermine and spermidine), that they may inhibit the uptake of the natural polyamines by the cells, and that, as a result, they deplete endogenous polyamine pools needed for cell replication. Cell death can result from any one of these effects or from several of them acting in tandem. On the other hand, the main advantage in the design of new tetramines is that structure-activity studies (SARs) have shown that relatively small structural changes in the aliphatic skeletons of the polyamines can cause pronounced differences in their pharmacological behavior and toxic side effects as well as in their antineoplastic activities, both at the cellular level as well as in animal models.
It has been repeatedly shown that 1N,14N-bisethylhomospermine (BE-4-4-4), a higher homologue of bisethylspermine, is a powerful cytotoxic drug but has a narrow therapeutic window. The results for animal trials reported against tumors such as L1210 leukemia or Lewis lung carcinoma grafted in athymic nude mice are indeed impressive; about a 6-fold increase in lifespan, as compared to control, was achieved. However, maximum tolerated dose (MTD) values for multiple injection (ip) schedules were found to be ca. 6 mg/kg, doses close to the levels necessary to achieve efficient antitumor activity in human tumor xenografts. It is therefore desirable to design compounds with a wider therapeutic window.
In pursuit of this goal, conformational restrictions were introducted into the BE-4-4-4 structure in order to increase its therapeutic activity, prompted by the promising results obtained with conformationally restricted analogues of bisethylspermine. Since the latter was found to be highly cytotoxic against a line of human prostate cancer cells, the new tetramines were assayed against several lines of human prostate cancer cells. LnCap, DU 145, DuPro, and PC-3. The results suggest that it is possible to markedly improve the therapeutic efficacy of BE-4-4-4-like compounds against human prostate cancer cells by introducing certain conformational restrictions.
Conformationally-restricted polyamine analogs and methods of synthesizing such analogs have been disclosed in U.S. Pat. Nos. 5,889,061 and 6,392,098 and International Patent Applications WO 98/17624 and WO 00/66587. In view of the utility of these compounds for treating neoplastic cell growth, additional synthetic methods are desirable, particularly synthetic methods amenable to large-scale synthesis. This invention provides improved methods for synthesizing cyclopropyl-containing polyamine analogs, as well as novel polyamine analogs containing cyclopropyl groups.